I have to measure the load data of my tool manually because I don’t have KUKA.LoadDataDetermination.
The meaning of M,X,Y,Z is clear, but what A, B, C mean? The manual reads: A, B, C: Orientation of the principal inertia axes relative to the flange. Does it related to the motion direction of my tool (mine is a mill spindle)?
And I am confused about JX, JY, JZ though it is said that JX is the inertia about the X axis of the coordinate system that is rotated relative to the flange by A, B and C.
Anyone helps to explain these parameters? Thanks!
Enter LoadData M XYZ ABC JXJYJZ
-
whitegreen -
April 28, 2015 at 10:33 AM -
Thread is marked as Resolved.
-
-
In case of Kuka, translations are labeled X,Y,Z and rotations are A,B,C where A is rotation about Z, B is rotation about Y etc.
The priority is A, then B, then C...
$FLANGE coordinate system is:
Z is out of flange
X is down (if A6 is at zero)
and Y is to the left (if A6 is at zero)When you specify X,Y,Z for load, you are using $FLANGE.
if A,B,C are zeroes, you are using $FLANGE so JX is inertia about flange axis X etc.
if A=90deg, B=0, C=0 then your principal axes are rotated so that X is to the left (where Y used to be) and Y is up (opposite where X used to be).
in case of inertia, sign does not matter (always positive). In this case JX is inertia when rotating about X axis but - X axis is now horizontal (if A6=0).Or here is perhaps a better example - You are responsible for robot programming, Kuka made the robot and someone else made tooling and gave you the load info in their drawing but .... axes don't match (see image).
You do the best to mount the tool using given bolt patterns and the tool can work but... mass was straight forward, after some pondering and visiting old geometry books you also figured out what to do with center of mass, the only problem is ... what do you do with inertia data? do you recompute it somehow? that will work but it is simpler to just transform the axes... (determine A,B and C).
for example, you could rotate about flange Z axis some neg. 60 deg and X axis will line up. at the same time, tooling Z axis will line up with flange Y. we could stop there and just enter JX as is, but during data entry swap values for JY and JZ. The other option is to do rotation about flange X axis by 90 deg - this will make all three axes line up....
-
Hi panic mode, thank you very much for the detailed replay! I’m glad that we are now talking about the very point: inertia measurement.
I understand ABC and axis rotation well for I have recently wrote java codes for the basic kinematics for KUKA. But I still have no idea how to measure (or caclucate) the tool inertia as you indicated in your illustration.
1. I have manually measured the mass (7.8kg), and the center of mass(x 49mm, y 0, z 66). The tool frame is along with the flange frame.
2. The inertia is calculated by regarding the tool as a solid box with evenly distributed weight. After checking some website on physics, I calculated JX=0.0078 kgm2, JX=JZ=0.0195 kgm2.
3. As you suggestedQuoterotate about flange Z axis some 60 deg and X axis will line up. at the same time, tooling Z axis will line up with flange Y. we could stop there and just enter JX as is, but during data entry swap values for JY and JZ.
, but then how to get the inertia data?
4. I don’t think rotating the tool (relative to flange) helps to get inertia data. Suppose a solid sphere tool centered at the flange center (though not physically possible), the inertia values do Not depend on the orientation at all, but on its radius and the density.Panic mode, maybe I have misinterpreted your idea of rotation?
-
there is tons of resources on this, check wiki.
simplest way is to use calculus and treat object as collection of point masses, then integrate.basic principle is
dJ = dm *r^2dm is small mass and r is distance from rotation axis.